Presentr Analyze

Translate and re-time subtitles, read text aloud, dub media and pull text from scans, with a quote you confirm first.

View on presentr.me

Hosted MCP Server

npx add-mcp 'https://analyze.presentr.me/api/mcp'

Installs into Claude Code, Codex, Cursor and more

Documentation

What it does

Presentr Analyze runs as a hosted MCP server, so the assistant or coding agent you already use can do the work for you: check and translate subtitles, re-time captions to their audio, read text aloud and dub audio or video. You sign in with your Presentr Analyze account in the browser; there's no key to copy.

Server address: https://analyze.presentr.me/api/mcp

Connect it

Claude Code

claude mcp add --transport http presentr-analyze https://analyze.presentr.me/api/mcp

Then run /mcp, choose presentr-analyze, then Authenticate, and approve in the browser.

Codex CLI

codex mcp add presentr-analyze --url https://analyze.presentr.me/api/mcp
codex mcp login presentr-analyze

Claude (claude.ai and the apps)

Add Presentr Analyze from its Claude directory listing and sign in when asked. The listing covers documents and subtitles. For read aloud and dubbing as well, add it as a custom connector instead: Settings → Connectors → Add custom connector, with the server address above.

ChatGPT

A ChatGPT listing is on its way. Until then, ChatGPT accounts that allow custom connectors can add ChatGPT's own address: https://analyze.presentr.me/api/mcp/chatgpt

VS Code, Cursor and other MCP clients

Add a remote (streamable HTTP) server with just the address:

{ "mcpServers": { "presentr-analyze": { "url": "https://analyze.presentr.me/api/mcp" } } }

We've tested browser sign-in with Claude Code. The server also accepts sign-in through a local (127.0.0.1 or localhost) redirect and the claude.ai, chatgpt.com and vscode.dev callbacks, which is what most clients use. If your client doesn't connect, tell us.

What you can ask

  • "Check this SRT's timing." — finds overlaps, cues that are too fast or too long, and gaps. Doesn't start a job.
  • "Translate these subtitles into Spanish." — returns the translated SRT and WebVTT.
  • "Re-time these subtitles to this audio." — aligns the captions to what's actually said.
  • "Read this text aloud." — an MP3, with a stock voice or one of your own voices.
  • "Dub this video into French." — the video (or audio) dubbed, plus the transcript and translated subtitles.

Local files: a coding agent can upload them for you (it's given a one-time upload link). Long media can also be passed as a direct link.

Every paid job starts with a quote

Before anything uses your credits, the agent has to get a quote: what the job will use, and your balance. Agents are instructed to show you the quote and start the job only once you confirm. A job never costs more than its quote, and if it fails you're not charged for what wasn't delivered. Credits are added on your account, never inside the agent.

For developers

The same work is available over the REST API with an API key. Agents that list tools see each one's inputs and outputs as JSON Schemas, and every result is structured JSON plus a short summary.